Bone Radiation-Induced Sarcomas: Outcomes Based on Histology and Surgical Treatment

Abstract

Background: Bone radiation-induced sarcomas (B-RIS) are secondary neoplasms with reportedly worse overall survival than de novo bone sarcoma. Treatment strategy for these neoplasms remains uncertain. Our systematic review sought to assess overall survival based on histology and surgical intervention. Methods: A systemic review was conducted following Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines and registered in PROSPERO (438415). Studies describing oncologic outcomes of patients with B-RIS in the appendicular and axial skeleton were included. The Strengthening the Reporting of Observational Studies in Epidemiology checklist was used for quality assessment. Survival analysis by histologic subtype and surgery type was performed in a subset of 234 patients from 11 articles with individualized data. A total of 20 articles with a total of 566 patients were included. The most frequent location was the pelvis (27.7%), and the main histological types were osteosarcoma (69.4%), undifferentiated pleomorphic sarcoma (14.1%), and fibrosarcoma (9.2%). Limb-salvage and amputation were performed in 68.5% and 31.5% of cases, respectively. Results: Local recurrence was 13%, without difference between limb-salvage surgery and amputation (p = 0.51). The metastasis rate was 42.3%. Five-year OS was 43.7% (95% confidence interval [CI], 33.3%-53.5%) for osteosarcoma, 31.5% (95% CI, 11.3%-54.2%) for UPS, and 28.1% (95% CI, 10.6%-48.8%) for fibrosarcoma. Five-year OS was 49.2% (95% CI, 35.3%-61.6%) for limb-salvage and 46.9% (95% CI, 29.1%-62.9%) for amputation. There was no difference in 5-year OS between histologic subtypes (p = 0.18) or treatment type (p = 0.86). Conclusion: B-RIS demonstrated poor OS at 5 years after initial management regardless of histology. Limb-salvage surgery was not associated with lower 5-year OS compared with amputation. Future studies should compare both groups while controlling for confounders. Level of Evidence: Level III. See Instructions for Authors for a complete description of levels of evidence.